Aim 1: We test the hypothesis that PRDM16 can drive the formation of functional brown adipocytes in white fat depots of mice to raise energy expenditure and protect against obesity and cold exposure. Diet-Induced Obesity (DIP). The first cohort of male aP2-PRDM16 and littermate control animals are currently in week 6 of a 16 week long study in which they are consuming a high fat containing diet (55% calories from fat). During the past year, out of necessity, we back-crossed the transgenic mouse lines into the C57Biack6 genetic background. The transgenic mice were generated in the FVB background and this inbred strain is not prone to high fat diet-induced obesity. We did not have any success generating transgenic lines directly in the C57Black6 strain. C57Black6 mice are the gold-standard for the types of metabolic experiments proposed. Cold-exposure. Acute and chronic cold exposure experiments will be performed in separate cohorts of transgenic and wildtype littermates in the C57Black6 genetic background. Transplantation of PRDM16 expressing fibroblasts. The goal is to develop an autologous transplantation protocol using PRDM16-expressing precursor cells to form ectopic deposits of functional brown adipose tissue (BAT) in animals. The hypothesis is that such "synthetic" BAT would safely and effectively raise energy expenditure to reduce obesity. In new data, PRDM16 expression in stromal vascular (SV) cells from white fat promoted adipogenesis in vivo after subcutaneous transplantion, however the transplants were always small (1-2 mm), had the unilocular appearance of white adipose and did not express the thermogenic marker, UCP1. In addition, skeletal myoblasts expressing PRDM16 formed ectopic depots of adipose tissue after subcutaneous transplantation. This result demonstrates that the adipogenic action of PRDIVI16 in committed myoblastic cells is compatible with the "real" environmental cues that control adipocyte differentiation in vivo. At the same time, mechanistic studies led by Shingo Kajimura demonstrated that PRDM16 binds and strongly coactivates c/EBPp (1). c/EBP(3 is well-expressed in myogenic precursor cells and is genetically required for PRDM16-driven brown adipogenic determination. Notably, BAT from c/EBPp-deficient fetal mice displayed dramatically reduced expression of thermogenic genes and elevated levels of muscle-specific transcripts analogous to what was observed in P/?OM) 6-deficient tissue. Co-expression of PRDM16 and c/EBPp in naive fibroblasts activates both the adipogenic and thermogenic gene program that are unique to brown adipocytes. Furthermore, these fibroblasts gave rise to multilocular, UCP1-expressing and ^^FDG-PET positive BAT in mice six weeks after subcutaneous implantation. These results indicate that a transcriptional unit containing c/EBPp and PRDM16 drives the brown adipose phenotype. Aim 2: We investigate the physiological requirement for PRDI\/I16 in BAT and whole body metabolism by creating and characterizing tissue-specific PRD/Wt6-deficient mice. We have now succeeded in generating 7 high percentage agouti chimeric mice from PRDMId"""^"^'* ES cell lines (4 independent cell clones). These animals are currently breeding with wildtype mice to establish germ-line transmission of the targeted allele. After the grant proposal was submitted, we became aware of PRD/W^6-deficient mice that were generated in the laboratory of David R. Beier at Brigham &Women's Hospital, Boston. The knock-out animals are perinatal lethal and have a pronounced cleft palate (B. Bjork, D. Beier et a/., in review). In collaboration with the Beier lab, we have analyzed the BAT phenotype from fetal PRD/W76-deficient mice. As now reported, Pf?DM^6-deficient BAT contains large lipid droplets and a dramatic reduction in the expression of thermogenic genes (2). This result reveals a genetic requirement for PRDM16 in the normal development of the tissue. In addition, the PRDM16^'BAT expressed higher levels of skeletal muscle (Sl\/l)-specific genes (2) supporting the notion that SIVl and BAT may arise from a common lineage (3). Although PRDA/f76-deficient BAT exhibits a dysregulated pattern of gene expression, it retains significant brown adipose attributes and is still recognizable as BAT. Therefore, the chronic loss of PRDM16 in vivo does not cause a total loss of BAT differentiation. This is different from the shRNA-based cell culture knockdown studies in which depletion of PRDIVlie in primary brown adipogenic precursors blocked brown adipogenesis and promoted overt skeletal myogenesis (2). We therefore hypothesize that another closely related member of the PR-domain containing family of proteins may partially compensate for the chronic loss of PRDM16 in BAT development. Indeed, PRDM3 and PRDM16 are closely related by sequence comparisons and share significant sequence similarity especially within the two conserved zinc-finger DNA/protein binding domains. We have therefore tested in our cell culture assays whether PRDM3 functions in brown adipogenesis. Preliminary studies show that PRDM3 is a key regulator of PPARY2 and adipogenic differentiation. Ectopic expression of PRDM3 in non-adipogenic fibroblasts such as NIH-3T3 cells stimulates adipogenesis including induction of some BAT-related genes in a cAMP dependent manner like UCP1 and PGC-1a. Aim 3: We investigate the mechanistic basis for PRDM16 function in: (1) activating adipogenesis via PPARy and (2) repressing myogenesis via binding to CtBP1/2.- Activation of PPARs. We have now reported that PRDIVI16 is a strong and ligand-dependent co-activator of both PPARy and PPARa (2). Moreover, agonism of PPARy was necessary for the adipogenic action of PRDM16 in cultured myoblasts. In vitro binding experiments revealed that PRDIVI16 is only able to bind to the full-length PPARy and does not bind to any of the isolated domains (N-terminal region, the ligand-binding domain or the C-terminus). This result suggests that PRDM16 makes crucial physical contacts in at least two domains of PPARy. PRDM16 binds via its two zinc finger regions to PPARy. Mass spectrometry analysis of PRDM16 transcriptional complexes showed that PRDM16 associates with p300/CBP in brown adipocyteswhether this acetyltransferase complex mediates activation of PPARs remains to be tested. Repression of Myogenesis. PRDM16 potently represses myogenic differentiation when ectopically expressed in C2C12 or primary satellite cell-derived myogenic cells (2). PRDM16 also associates with a known repressor complex containing C-terminal binding protein-1 or -2 (CtBP-1, 2) to suppress white-fat gene expression (4). The requirement for CtBP in the PRDM16-dependent repression of muscle specific genes was examined in C2C12 myoblasts using a mutant form of PRDM16 that no longer associates with CtBPs but retains its capacity to induce brown adipogenic genes (4). Interestingly, the CtBP binding mutant form of PRDM16, PRDMie''^'was unable to repress some muscle specific genes (e.g. myogenin) but retained its repressive effect on others (e.g. MyoD). This result suggests that a PRDM16/CtBP complex mediates the repression of some but not all muscle-related genes. To define other factors that may contribute to the repression of muscle genes by PRDM16, we performed a mass spectrometry analysis of the PRDM16 complex in C2C12 cells. A number of candidate proteins that may be involved in the PRDM16-driven cell fate transition between myoblasts and brown fat cells were identified.